To determine whether the anti-endotoxin human monoclonal antibody (mab-T88) given as a single IV infusion to patients with gram-negative sepsis can decrease 28 day mortality from sepsis, as compared to a placebo control.